Neural network model for predicting peak photochemical pollutant levels

In this paper, an attempt is made for the 24-hr prediction of photochemical pollutant levels using a neural network model. For this purpose, a model i s developed that relates peak pollutant concentrations to meteorological an d emission variables and indexes. The analysis is based on measurements of O-3 and NO2 from the city of Athens. The meteorological variables are selec ted to cover atmospheric processes that determine the fate of the airborne pollutants while special care is taken to ensure the availability of the re quired input data from routine observations or forecasts. The comparison be tween model predictions and actual observations shows a good agreement. In addition, a series of sensitivity tests is performed in order to evaluate t he sensitivity of the model to the uncertainty in meteorological variables, Model forecasts are generally rather insensitive to small perturbations in most of the input meteorological data, while they are relatively more sens itive in changes in wind speed and direction.